Construction and application of artificial intelligence crowdsourcing map based on multi-track GPS data

Yong Wang,Yanlin Zhou,Huan Ji,Zheng He,Xinyu Shen

2024-02-24

Abstract:In recent years, the rapid development of high-precision map technology combined with artificial intelligence has ushered in a new development opportunity in the field of intelligent vehicles. High-precision map technology is an important guarantee for intelligent vehicles to achieve autonomous driving. However, due to the lack of research on high-precision map technology, it is difficult to rationally use this technology in the field of intelligent vehicles. Therefore, relevant researchers studied a fast and effective algorithm to generate high-precision GPS data from a large number of low-precision GPS trajectory data fusion, and generated several key data points to simplify the description of GPS trajectory, and realized the "crowdsourced update" model based on a large number of social vehicles for map data collection came into being. This kind of algorithm has the important significance to improve the data accuracy, reduce the measurement cost and reduce the data storage space. On this basis, this paper analyzes the implementation form of crowdsourcing map, so as to improve the various information data in the high-precision map according to the actual situation, and promote the high-precision map can be reasonably applied to the intelligent car.

Artificial Intelligence,Human-Computer Interaction

What problem does this paper attempt to address?

The problem this paper attempts to address is: how to generate high-precision GPS data by integrating a large amount of low-precision GPS trajectory data, and construct an AI-based crowdsourced map based on multi-trajectory GPS data to improve the positioning accuracy and data update efficiency of autonomous vehicles.

Specifically, the paper focuses on the following aspects:
1. **Application of high-precision map technology**: High-precision map technology is an important guarantee for the realization of autonomous driving, but there is currently a lack of effective research to reasonably utilize this technology.
2. **Data fusion algorithm**: A fast and effective algorithm is proposed to generate high-precision GPS data points from a large amount of low-precision GPS trajectory data, simplifying the description of GPS trajectories.
3. **Crowdsourced update model**: Through large-scale data collection by social vehicles, the "crowdsourced update" of map data is realized, improving the accuracy and real-time performance of the data.
4. **Data processing and optimization**: Through mathematical modeling and multi-GPS trajectory data fusion algorithms, the data processing process is optimized, reducing measurement costs and storage space.

These solutions aim to improve the data quality of high-precision maps, reduce operating costs, and enhance the safety and efficiency of autonomous vehicles.